At sites of chronic inflammation induced during the course of a variety of infections, autoimmune diseases, allograft rejection, graft-versus-host disease and immunotherapy of cancer using interleukin-2 (IL-2), significant endothelial cell injury is known to occur, leading to pathogenesis of severe toxicity, the underlying reason for which remains unclear. Studies from this lab demonstrated that a variety of activated cytotoxic lymphocytes [cytotoxic T lymphocytes (CTL), lymphokine-activated killer (LAK) cells, and double negative (DN) T cells] can mediate efficient MHC-unrestricted lysis of target cells, including the endothelial cells, when activated through CD44. Also transfer of CTL+IL-2 into irradiated mice could trigger endothelial cell injury and vascular leak syndrome (VLS). Moreover, they have also shown that perforin knockout (KO) and FasL-defective mice exhibit decrease IL-2-induced VLS. The current study is to delineate the mechanism of endothelial cell injury by using IL-2-induced VLS as a model. It will test the hypothesis that dysregulation of the interaction between activated CTL or LAK cells expressing CD44 and endothelial cells bearing the appropriate ligand such as hyaluronic acid (HA), could lead to endothelial cell lysis and induction of VLS. The role of CD44 in the induction of VLS will be investigated using CD44 KO mice. The question of whether such mice are resistant to VLS induction upon IL-2 administration and whether IL-2 activated CTL or LAK cells from these mice can trigger VLS will be addressed. The role played by CD44 variant isoforms in adhesion and cytotoxicity of endothelial cells will be studies using CD44-exon specific KO mice and transfection of isoform-specific constructs in CD44-deficient cytolytic lymphocytes. To further test the hypothesis that VLS results from direct endothelial cell lysis, the role of FasL, perforin and TNF in VLS induction will be studied using double-KO (FasL and perforin-deficient) and TNFR-KO mice. Lastly, based on the above results, attempts will be made to block the VLS induction in vivo using soluble CD44Rg fusion proteins and mAbs against CD44 variant isoforms. The role of CD44-hyaluraonate interactions in endothelial cell injury will be tested using a mutant fusion protein (CD44MutRg) or Fab fragments of anti-CD44 mAbs. Together, the studies should shed new lights on the basic mechanisms involved in cytotoxic-endothelial cell interactions leading to endothelial cell injury and possible therapeutic approaches to prevent this, in a variety of disease states, including immunotherapy of cancer.